Uniparental Propagation of Mitochondrial DNA in Mouse-Human Cell Hybrids

Home , Chromosome 10, Chromosome 11, Chromosome 13, Chromosome 14, Chromosome 15, Chromosome 17

Proc. Natl. Acad. Sci. USA Vol. 77, No. 7, pp. 4079-4083, July 1980 Cell Biology Uniparental propagation of mitochondrial DNA in mouse-human cell hybrids (chromosomes/isozymes/restriction enzyme/Southern hybridization) LAURA DE FRANCESCO*, GIUSEPPE ATTARDI*t, AND CARLO M. CROCE* *Division of Biology, California Institute of Technology, Pasadena, California 91125; and tThe Wistar Institute of Anatomy and Biology, Philadelphia, Pennsylvania 19104 Communicated by Hilary Koprowski, March 13,1980

ABSTRACT The retention of the two parental mitochon- thymidine (8). Hybrids in group 2 were formed between the drial DNAs has been investigated in a large number of cell line HT-1080 and the contact-inhibited continuous mouse mouse-human cell hybrids segregating either mouse or human cell line THO-2, which is a ouabain-resistant, hypoxanthine chromosomes, by using a highly sensitive and specific method 3T3 derivative (9), and for detection of the DNA; the results have been correlated with phosphoribosyltransferase-deficient the karyotype and isozyme marker pattern in the same hybrid were selected in hypoxanthine/aminopterin/thymidine sup- lines. In e hybrids examined, a consistent pattern was ob- plemented with ouabain. Hybrids in group 3 were formed served for the type of mitochondrial DNA retained: the mito- between an a-amanatin-resistant variant of the human cell line chondrial DNA of the parent whose chromosomes were segre- HT-1080-6TG and the mouse L cell derivative clone ID, and gated from the nucleus was undetectable or present in marginal were selected in hypoxanthine/aminopterin/thymidine con- amounts. This was true also of hybrids containing a complete taining 7 ,ug of a-amanitin per ml (Boehringer Mannheim). set of the segregating chromosomes in the total or a large frac- Karyologic Analysis. Metaphase chromosomes of hybrid tion of the cell population. cells were banded according to a modification of the trypsin/ Previous investigations using hybrid cells derived from fusion Giemsa method of Seabright (10, 11). of cells from mice and humans (1-3), rats and humans (3), or Isozyme Analysis. M>H hybrid cells were studied for the mice and hamsters (4) have established in these hybrids a cor- expression of isozyme markers assigned to each of the different relation, though not very strong, between chromosome and human chromosomes by starch gel or cellulose acetate gel mitochondrial DNA (mtDNA) segregation. However, in none electrophoresis (12, 13): peptidase C (PEP-C), adenylate kinase of these studies has a detailed karyological analysis of the hy- 2 (AK-2), phosphoglucomutase 1 (PGM-1), and enolase 1 brids been performed to establish the rules that govern the re- (ENO-1) on chromosome 1; acid phosphatase 1 (ACP-1) and tention of the two parental mtDNAs in interspecific cell hy- isocitrate dehydrogenase (IDH) on chromosome 2; /3-galac- brids. Thus, it is still not known whether the disappearance of tosidase ((3-GAL) on chromosome 3; phosphoglucomutase 2 mtDNA of one parental species in the hybrid cell depends on (PGM-2) on chromosome 4; hexosaminidase B (HEX-B) on the loss of one chromosome or set of chromosomes of that species chromosome 5; malic enzyme (ME), phosphoglucomutase 3 or on an imbalance of chromosomes of the two species or on a (PGM-3), glyoxalase 1 (GLO-1), and superoxide dismutase 2 more complex regulatory phenomenon. (SOD-2) on chromosome 6; mitochorknrial malic dehydrogenase A large series of hybrids between the human cell line HT- (mMDH) and fl-glucuronidase (,8-GUS) on chromosome 7; 1080 and mouse cells has been isolated (5). These hybrids tend glutathione reductase (GTR) on chromosome 8; adenylate ki- to lose chromosomes of either species depending on the mouse nase 1 and 3 (AK-1 and -3) and mitochondrial aconitase parent, but many of them retain in a relatively stable form a (mACO) on chromosome 9; glutamic oxaloacetic transaminase large number of chromosomes of both species. In this paper we (GOT) on chromosome 10; lactic dehydrogenase A (LDH-A) present the results of a parallel systematic investigation of the and acid phosphatase 2 (ACP-2) on chromosome 11; lactic de- mtDNA composition, karyotype, and isozyme marker pattern hydrogenase B (LDH-B) and peptidase B (PEP-B) on chro- in a fairly large number of these hybrids. By using a highly mosome 12; esterase D (EST-D) on chromosome 13; nucleoside sensitive and specific method for detection of mtDNA of the phosphorylase (NP) on chromosome 14; mannose phosphate two parental species, we show that, in all hybrids analyzed, the isomerase (MPI), pyruvate kinase 3 (PK-3), and a subunit of mtDNA of the species being segregated from the nucleus was hexosaminidase A (HEX-A) on chromosome 15; adenine undetectable or present only in minute amounts. phosphoribosyltransferase (APRT) on chromosome 16; thymidine kinase (TK) and galactokinase (GALK) on chromosome METHODS 17; peptidase A (PEP-A) on chromosome 18; glucose phosphate Cell Hybrids Used. The methods for production and selec- isomerase (PGI) on chromosome 19; adenosine deaminase tion of mouse-human hybrid cell lines that lose chromosomes (ADA) on chromosome 20; superoxide dismutase 1 (SOD-1) on of either species have been described (5-7). A complete list of chromosome 21; arylsulfatase A (ARS-A) on chromosome 22; all hybrids analyzed in this study and the parental cell lines of and glucose-6-phosphate dehydrogenase (G6PD), hypoxanthine each are presented in Table 1. Hybrids in group 1 were formed phosphoribosyltransferase (HPRT), and phosphoglycerate ki- between a thioguanine-resistant variant (6TG) of the human nase (PGK) on the X chromosome. cell line HT-1080 (5) and either mouse peritoneal macrophages H>M hybrid cells were studied for the expression of isozyme or cells derived directly from the solid mouse teratocarcinoma markers assigned to the following mouse chromosomes: di- and were selected in peptidase 1 (DIP-1) on chromosome 1; adenylate kinase 1 OTT-6050 hypoxanthine/aminopterin/ (AK-1) on chromosome 2; carbonic anhydrase (CA) on chro- enolase 1 on chromosome 4; ,B-glucuron- The publication costs of this article were defrayed in part by page mosome 3; (ENO-1) charge payment. This article must therefore be hereby marked "ad- vertisement" in accordance with 18 U. S. C. §1734 solely to indicate Abbreviation: mtDNA, mitochondrial DNA. this fact. t To whom reprint requests should be addressed. 4079 Downloaded by guest on September 24, 2021 4080 Cell Biology: De Francesco et al. Proc. Natl. Acad. Sci. USA 77 (1980)

Table 1. Hybrid cell lines analyzeed 1. Hybrids losing mouse chromosomes (H>M) 9 HT-1080-6TG X BALB/c MPM* t55-14-Fl 9 t55-14-Fl Cl 29 9 t55-14-F7 9 t55-54-F4 I 9.55-91-F2 Cl 4 I t55-91-F2 Cl 15 HT-1080-6TG x 129 MPM III C11-7,Cl1-13 Cl 1-15, Cl 1-16 Cl 1-17, Cl 1-20 FIG. 1. Restriction enzyme HT-1080-6TG X OTT-6050 55-84-F8 patterns of human (Right) and 2. Hybrids losing human chromosomes (Mct > H) mouse (Left) mtDNA. The auto- HT-1080 X THO-2 56-05-F4 Cl 6 -radiograph shows the electropho- ~~~~~~reticseparation on a 1.4% agarose 56-05-F4 Cl 16 ... gel of the products obtained by 56-05-F5 Cl 7 FHincG1digestion of mtDNA from 56-05-F5 Cl 10 human cells and mouse liver cells. 3. Hybrids losing mouse chromosomes (H > Mct) The fragments had been labeled HT-1080-6TG, a-amR X Cl1iD 58-92-Fl Cl 4 with Escherichia coli DNA poly- 58-92-F2 Cl 5 merase I and deoxynucleoside 58-92-F3 Cl 10 [a-32p~triphosphates (16). * MPM, mouse peritoneal macrophages. t Mc, mouse continuous cell line. The digest was divided into two equal portions which were run idase ((-GUS) on chromosome 5; triphosphate isomerase (TP-1) in separate lanes on a 1.4% agarose gel in Tris acetate buffer (10 on chromosome 6, lactic dehydrogenase A (LDH-A) on chro- mM Tris, pH 7.4/50 mM sodium acetate/2.5 mM EDTA) in mosome 7; glutathione reductase (GR) on chromosome 8; parallel with HincII-digested mouse and human mtDNAs. The mannose phosphate isomerase (MPI) on chromosome 9; tri- gels were blotted onto nitrocellulose paper by the Southern peptidase (TRIP) on chromosome 10; galactokinase (GALK) technique (17). Separate filters, each containing the hybrid on chromosome 11; acid phosphatase 1 (ACP-1) on chromosome mtDNA and the mouse and human mtDNA standards, were 12; nucleoside phosphorylase (NP) on chromosome 14; glutamic incubated with 2 X 106 cpm (w20 ng) of each parental mtDNA pyruvic transaminase (GPT) on chromosome 15; glyoxalase probe prepared essentially as described by Rigby et al. (18) in (GLO) on chromosome 17; dipeptidase 2 (DIP-2) on chromo- 4 ml of 0.9 M NaCl/0.09 M sodium citrate/0.1% NaDodSO4/ some 18; glutamic oxaloacetic transaminase (GOT) on chro- 0.02% bovine serum albumin/0.02% Ficoll at 680C for 20 hr mosome 19; and glucose-6-phosphate dehydrogenase (G6PD) in Seal-a-Meal bags (Daisey Products, Industrial Airport, K). on the X chromosome. After hybridization, the filters were washed for 10-15 min at Cell Growth and Labeling Conditions. Cells were main- 680C in hybridization buffer without probe, then for 10-15 min tained as monolayers in Dulbecco's modified Eagle's medium in 75 mM NaCl/7.5 mM sodium citrate/0.1% NaDodSO4 at the plus 10% fetal calf serum. To provide a marker for following same temperature, and finally exposedtop an x-rayfilm with the DNA during purification and estimating the quantity of one screen intensifier at -700C. mtDNA recovered, cells to be used for mtDNA isolation were labeled for 2-3 days in the presence of [3H]thymidine (0.2 RESULTS MCi/ml; 1 Ci = 3.7 X 1010 becquerels). Analysis of mtDNA. The approach used in this work took Isolation of mtDNA. A mitochondrial fraction was prepared advantage of both the distinctive restriction pattern of the from 10-20 plates of each hybrid cell line by differential cen- mtDNA of each species and the base sequence differences be- trifugation (14). Total mtDNA was isolated by a published tween human and mouse mtDNAs (3,19), increasing, therefore, procedure (15). Closed circular mtDNA was isolated as de- the specificity of recognition of the origin of the mtDNA in the scribed from HeLa cells (16) or from the livers and kidneys of cell hybrids. Furthermore, it was possible, by this approach, to mice. recognize directly the presence of any recombinant molecules Restriction Digestion, Southern Blots, and DNA-DNA as giving rise to fragments reacting with both parental probes. Hybridization. Approximately 100 ng of hybrid cell mtDNA Finally, because the probes were labeled in vitro to a much was digested with the restriction enzyme HincIH (BioLabs) higher specific activity (in the present work the specific activity under standard conditions with a 4- to 6-fold excess of enzyme. varied between 2 and 8 X 107 cpm/(.1g) than obtainable in in

Table 2. Mouse chromosomes in H>M cell hybrids Mouse chromosome* Hybrid 1' 2 3 4 5 6 7 8 9 10 11. 12 13 14 15 16 17 18 19 X Markers 55-14-F7 1.0 0.7 1.1 0.4 0.8 1.0 0.8 0.8 1.2 1.6 0.7 0.3 0.6 0.3 0.8 0.8 0.9 1.3 1.3 0.6 1.1 55-14-Fl C129 1.0 0.8 0.6 0.5 0.9 0.4 0.9 1.2 0.5 0.5 1.1 0.5 0.1 0.1 0.1 0.3 0.8 0.8 1.0 0.05 5.2 IIIC1 1-7 0.9 1.2 0.5 0.4 1.0 1.8 0.8 1.6 1.0 2.0 1.3 1.2 1.0 0.9 0.8 0.4 1.1 1.2 1.0 0.5 3.5 IIIC1 1-15 1.2 0.9 0.9 0.5 1.2 1.1 1.1 1.3 0.8 1.0 0.6 0.6 0.5 0.4 0.2 0.5 0.3 0.2 0.3 0.1 2.0 55-84-F8 1.7 1.7 0.8 0.3 1.7 2.0 1.1 2.4 1.6 2.0 1.3 1.0 1.2 1.2 1.0 1.2 1.6 2.2 2.0 0.9 8.0 55-14-Fl 0.4 0.4 1.1 0.7 0.85 1.1 0.85 0.7 0.4 0.8 0 0.8 0.6 0.4 0 0 0.3 0.3 0.6 0.14 4.6 55-54-F4 1.5 0.8 1.5 1.5 0.8 0.9 0.8 1.3 0.9 0.6 0.4 1.0 0.2 0.4 0.3 0.2 0.4 0.5 0.6 0.4 1.2 55-91-F2 C0 4 1.1 0.6 1.7 1.3 0.6 0.8 0.6 1.0 0.45 0.5 0 0.4 0 0.5 0 0 0.1 0.2 0.45 0.45 0.6 55-91-F2 Cl 15 0.7 0.5 0.7 0.4 1.1 0.4 0.3 1.0 0.3 0.2 0.2 0.05 0.1 0.2 0.1 0.4 0.4 0.7 1.6 0.4 6.0 * Values given are the fraction of each mouse chromosome present per cell. Downloaded by guest on September 24, 2021 Cell Biology: De Francesco et al. Proc. Nati. Acad. Sci. USA 77 (1980) 4081

Table 3. Expression of mouse isozymes in H>M cell hybrids Isozymes DIP-1 AK-1 CA ENO-1 A-GUS TP-1 LDH-A GR MPI TRIP GALK ACP-1 NP GPT GLO DIP-2 GOT G6PD Hybrid (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (14) (15) (17) (18) (19) (X) 55-14-F7 + + + + + + + + + + + + + + + + + + 55-14-Fl Cl29 + + + + + + + + + + + + Weak Weak + + + Weak IIICl1-7 + + + + + + + + + + + + + + + + + + III Cl 1-15 + + + + + + + + + + + + + + + + + Weak 55-84-F8 + + + + + + + + + + + + + + + + + + 55-14-Fi + + + + + + + + + + - + + - + + + Weak 55-54-F4 + + + + + + + + + + + + + Weak + + + + 55-91-F2 Cl 4 + + + + + + + + + + - + + - Weak + + 55-91-F2 Cl 15 + + + + + + + + + Weak Weak - + Weak + + + + Abbreviations used for the isozymes are indicated in Methods. Numbers in parentheses are chromosome numbers.

vvo labeled DNA, the sensitivity of detection of minor com- mtDNA in Cell Hybrids Losing Mouse Chromosomes ponents was greatly increased. Consequently, much smaller (H>M). The human cell line HT-1080-6TG fused to mouse amounts of cells were sufficient for study than otherwise re- peritoneal macrophages produced hybrid cells that tended to quired for restriction enzyme analysis of in vivo labeled DNA. lose mouse chromosomes and that retained the entire comple- This advantage is particularly important to avoid the changes ment of human chromosomes (5). Several hybrids of this type in the chromosome constitution of the hybrid cell population (Table 1, group 1), were analyzed for the nuclear chromosome which occur during prolonged growth. complement and for the species of mtDNA retained. Both The restriction enzyme chosen for this analysis was HlncII, karyotypic (Table 2) and isozyme (Table 3) analyses were done, which makes 11 cuts in human mtDNA (D. Ojala, X. Shaffer, and by both criteria, it can be estimated that a large number and X. Baskir, personal communication) and 5 cuts in mouse of mouse chromosomes were present in the majority of the cells. mtDNA (19), generating fragments from both mtDNAs that Representative examples of the blots of the mtDNA of these are easily separated and recognized on gels (Fig. 1). Early in hybrids are shown in Figs. 2 and 3. The mtDNA from hybrid this work, we found that the nick-translated mtDNA probes 55-14F7, which was derived from the fusion of HT-1080 cells reacted extensively with the heterologous mtDNA. However, to macrophages from BALB/c mice, showed a normal human by including a final wash in low salt at high temperature, (75 pattern when probed with human mtDNA (Fig. 2 Left). In this mM NaCl/7.5 mM sodium citrate; 680C) of the blot after hy- blot, it is possible to see some faint bands in the mouse standard bridization, it was possible in general to eliminate most of the lane, which correspond to mouse mtDNA fragments cross- cross-hybridization, except that involving the rRNA gene region hybridizing slightly with the human probe. When the hybrid (20). In the human pattern, the rRNA genes are located in mtDNA was probed with mouse mtDNA, several faint bands HincH fragments 1, 6, and 7 (D. Ojala, J. Shaffer, and B. Baskir, were observed that comigrated with bands seen in the human personal communication) and, in the mouse pattern, predom- standard and, therefore, presumably represent cross-hybri- inantly in fragments B and D (19). dizing human mtDNA fragments (Fig. 2 Center). In addition, there was a strong band that corresponded roughly to the position of uncut mtDNA (arrow in Fig. 2 Center). In order to investigate the identity of this band, we prepared nick-translated probes of mouse standard mtDNA from either the ma- terial corresponding to the slowly migrating band or from the combined four HincII mouse mtDNA fragments, A-D. The results of the hybridization of the latter probe with the mtDNA

I @ - 1, --

WI _ i I. . I, ": :.o :- }1 Tl iif, 4 \ I

* ,.. w A j 0- . _ -

M -- doZ'maim--...... _ X7 _

*5 _m-- i, -

.._ , d_ i_lw ;t- .A FIG. 2. Restriction fragment analysis of mntDNA from a hybrid between human HT-1080-6TG cells and BALB/c mouse macrophages, segregating mouse chromosomes (H>M). Samples of HincII-digested mtDNA from the hybrid 55-14-F7 (Hy) were run in parallel with equivalent amounts of HincII-digested mouse (Mou) and human (Hu) mtDNA, blotted onto nitrocellulose paper, and FIG. 3. Restriction fragment analysis of mtDNA from hybrids hybridized with 2 X 106 cpm (--20 ng) of nick-translated human (55-14-Fl Cl 29, m Cl 1-7, and III Cl 1-15) (Hy) between human mtDNA (Left), mouse mtDNA (Center), and mouse mtDNA Hincd HT-1080-6TG cells and BALB/c or 129 mouse macrophages, segre- fragments A-D (Right). gating mouse chromosomes (H>M). Downloaded by guest on September 24, 2021 4082 Cell Biology: De Francesco et al. Proc. Natl. Acad. Sci. USA 77 (1980) Table 4. Expression of human isozymes in Mc>H cell hybrids In Fig. 3 Left, hybrid 55-14-Fl Cl 29, which has the same Human Human Mc>H hybrids (THO-2 X HT-1080) parents as the previous hybrid but lacks a larger number of chromo- isozyme 56-05-F4 56-05-F5 56-05-F5 56-05-F4 mouse chromosomes (Tables 2 and 3), is shown to have a normal some marker C0 6 C0 10 C0 7 C0 16 human mtDNA pattern when probed with human mtDNA. In 1 PEP-C, AK-2, the blot incubated with mouse mtDNA probe, it is possible to PGM-1, ENO-1 - - + + see very faint mouse mtDNA bands (arrows), which probably 2 ACP-1, IDH - - + + represent less than 1% of the total mtDNA in the hybrid, in 3 a-GAL + + + + addition to the more prominent cross-hybridizing human 4 PGM-2 + + + + mtDNA fragments. In Fig. 3 Right are shown two other hybrids 5 HEX-B - - Weak+ - formed by the fusion of HT-1080-6TG and mouse macrophages 6 ME, PGM-3, from strain 129. Even upon very long exposure of the autora- GLO-1, SOD-2 - - + + diogram, there were no mouse mtDNA fragments visible. 7 mMDH, ,-GUS + + + When HT-1080-6TG cells were fused with mouse terato- 8 GTR + + + - carcinoma cells (OTT-6050), hybrid cells losing mouse chro- 9 AK-1, AK-3, mosomes wete also obtained (6). A single hybrid of this type was mACO + + - + analyzed (5,-84-F8) (the chromosome and isozyme patterns 10 GOT + + + + are shown in Tables 2 and 3, respectively). Here again, a normal 11 LDH-A, ACP-2 + + + + human mtDNA pattern was seen with the human mtDNA 12 LDH-B, PEP-B + + + + probe, and no mouse mtDNA fragments were detected in the 13 EST-D - + Weak+ + hybridization with the mouse mtDNA probe. 14 NP + + + + mtDNA in Hybrids Losing Human Chromosomes (Mc>H) 15 MPI, PK-3, When human HT-1080 cells were fused with the continuous HEX-A + + Weak+ + mouse cell line THO-2, hybrids were obtained that preferen- 16 APRT + + + + tially lost human chromosomes upon continuous culture, but 17 TK, GALK + + Weak + + many of the hybrids retained large numbers of human chro- 18 PEP-A + - - + mosomes In 19 PGI + + + + (5). the hybrids investigated here (Table 1), from 20 ADA + + + - the analysis of the isozymes present it would be inferred that 21 SOD-1 + + + + between 16 and 19 of the human chromosomes were present 22 ARS-A + - - - in the majority of the cells (Table 4). As shown in Fig. 4 Left, X G6PD, HPRT, hybrid 56-05-F4 Cl 6 revealed a normal mouse mtDNA pattern; PGK + Weak+ + + even upon long exposure of the autoradiogram, the blot hybridized with human mtDNA showed no human mtDNA Abbreviations used for the isozymes are indicated in Methods. fragments. The faint band with the mobility of human fragment 7 is presumably a contaminant because it is also present of the hybrid and with the standard DNAs are shown in Fig. in the mouse standard. Hybrid 56-05-F5 Cl 10 gave a similar 2 Right. With this probe, hybridization to the slowly migrating result (Fig. 4 Right). The faint bands visible between the A and component has been nearly completely eliminated, indicating B and between the B and C mouse fragments are not human that this component probably does not contain mouse mtDNA sequences. In support of this conclusion, when the probe prepared from the slowly moving component was annealed with , HincII-digested mouse mtDNA, no hybridization was observed .* to mouse fragments A-E, but a strong band of hybridization was observed at the position of the large component (not shown). The band with the mobility of mouse mtDNA fragment ..d^. 14 F

B appearing in the restriction pattern of the F7 mtDNA * hy- I - 3iA iie* # bridized with human probe (Fig. 2 Left) or mouse probe (Fig. 2 Right) probably represents a human mtDNA partial, because A it is present also in the human standard and because no bands corresponding to the other mouse mtDNA fragments are observed (the same partial is visible in the blots in Fig. 6).

1_.__. .4 'Al

FIG. 5. Karyotype ofa hybrid clone between a-amanitin-ristnt HT-1080-6TG human fibrosarcoma cells and Cl 1D mouse cells. Al- FIG. 4. Restriction fragment analysis of mtDNA from hybrids most all ofthe mouse Cl 1D chromosomes (Cl 1D cells have approxi- (56-05-F4 C16 and 56-05-F5 Cl 10) (Hy) between human HT-1080 mately 50 mouse chromosomes) are present in the hybrid. Six small cells and the continuous mouse cell line THO-2, segregating human marker chromosomes of unknown origin (shown at the end of the chromosomes (Mc>H). mouse set) are present in the hybrids. Downloaded by guest on September 24, 2021 Cell Biology: De Francesco et al. Proc. Nati. Acad. Sca. USA 77 (1980) 4083

.. controlled by the unstable parent are suppressed (23). It is clear, .. '. however, that suppression of these functions does not entail the inactivation of the entire genome, because marker isozymes of the suppressed parent, which are coded for by genes distributed on nearly all chromosomes, can be detected in hybrid cells. The nature of the suppression at the molecular level is unknown. 404P The results obtained with the cell hybrids analyzed here agree with previous studies in which the retention of a single species of mtDNA was demonstrated (1, 2). However, in the -4m studies on rodent-human hybrid cells performed by Coon et al. (3), several cell hybrids were found in which both species of parental mtDNA were detected. The reason for this dis- crepancy is not clear. It seems likely that the mouse-human cell hybrids are in all cases unable to replicate both species of mtDNA indefinitely and that the hybrids analyzed in the cited study had not yet reached the point of complete elimination of one species of mtDNA. It would be important to establish FIG. 6. Restriction fragment analysis of mtDNA from hybrids whether hybrid cells are indeed capable of propagating mtDNA (58-92-Fl C1 4, 58-92-F2 Cl 5, and 58-92-F3 Cl 10) (Hy) between even for a limited time, as this would suggest a different a-amanitin-resistant human HT-1080-6TG cells and the continuous mechanism of suppression of the genes required for mtDNA mouse cell line clone iD, segregating mouse chromosomes retention from that operating on the rRNA genes; the inacti- (H>Mc). vation of rRNA genes has, in fact, been shown to occur within very few cell generations after nuclear fusion (24). mtDNA fragments, but rather either contaminants or mouse mtDNA partials; they are not visible after hybridization with This work was supported by Research Grant PCM-7718037 from the National Science Foundation and a Junior Fellowship of the the human probe and no other human fragments are visible. American Cancer Society, California Division (to L.D.), and by grants mtDNA in Hybrids Formed Between a-Amanitin-Resis- from the National Institutes of Health and the National Foundation- tant Human Cells and Mouse LM (U-) Cl ID Cells (H>Mc) March of Dimes. In all hybrids derived from the fusion of established lines of mouse and human cells, the human chromosomal complement 1. Clayton, D. A., Teplitz, R. L., Nabholz, U., Dovey, H. & Bodmer, is unstable and human chromosomes tend to be shed during W. (1971) Nature (London) 234,560-562. 2. Attardi, B. & Attardi, G. (1972) Proc. Natl. Acad. Sci. USA 69, continued culturing. However, when a-amanitin-resistant 2874-2878. human HT-1080-6TG cells were fused with mouse clone iD 3. Coon, H. G., Horak, I. & Dawid, I. B. (1973) J. Mol. Blol. 81, cells and hybrids selected in the presence of a-amanitin, the 285-298. entire chromosomal complement of human origin was retained 4. Eliceiri, G. L. (1973) Nature (London) New Biol. 241, 233- in an apparent stable state and the mouse chromosomes showed 234. 5. Croce, C. M. (1976) Proc. Natl. Acad. Sct. USA 73, 3248- only a very slow loss; these hybrids expressed only human 28S 3252. rRNA (M. Shander and C. Croce, unpublished data). Several 6. Miller, 0. J., Miller, D. A., Dev, V. G., Trantravahi, R. & Croce, hybrids of this kind (Fig. 5), which had a complete human and C. M. (1976) Proc. Natl. Acad. Sci. USA 73,4531-4535. mouse chromosome complement in the majority of the cells, 7. Croce, C. M., Talavera, A., Basilico, C. & Miller, 0. J. (1977) Proc. were analyzed for the species of mtDNA retained; only human Nati. Acad. Sci. USA 74,694-697. 8. Littlefield, J. W. (1964) Science 145,709-710. mtDNA was detected (Fig. 6). 9. Jha, K. K. & Ozer, H. L. (1976) Somatic Cell Genet. 2, 215- 223. DISCUSSION 10. Seabright, M. (1971) Lancet ii, 971-972. 11. Croce, C. M., Kieba, L. & Koprowski, H. (1973) Exp. Cell Res. 79, In all cell hybrids analyzed in the present work, regardless of 461-463. the chromosome composition and direction of chromosome loss, 12. Chern, C. & Croce, C. M. (1976) Am. J. Hum. Genet. 28, a single species of mtDNA was detected, and this belonged to 350-356. the parent whose nuclear chromosomes were more stable. In 13. O'Brien, D., Linnenbach, A. & Croce, C. M. (1978) Cytogenet. one set of hybrids (formed between a-amanitin-resistant human Cell Genet. 21, 72-76. 14. Attardi, B., Cravioto, B. & Attardi, G. (1969) J. Mol. Biol. 43, cells and a-amanitin-sensitive mouse cells), there was at least 47-117. one copy of each mouse chromosome, in addition to several 15. Storrie, B. & Attardi, G. (1972) J. Mol. Biol. 71, 177-199. copies of each human chromosome, in most hybrid cells. This 16. Ojala, D. & Attardi, G. (1977) Plasmid 1, 78-105. result makes it unlikely that the absence of a specific chromo- 17. Southern, E. M. (1975) J. Mol. Biol. 98,503-517. some or set of chromosomes of one parent is responsible for the 18. Rigby, P. W. J., Dieckmann, M., Rhodes, C. & Berg, P. (1977) of the cells to and maintain J. Mol. Biol. 113,237-251. inability hybrid replicate the 19. Parker, R. C. & Watson, R. M. (1977) Nucleic Acids Res. 4, mtDNA of that parent and is suggestive of some kind of regu- 1291-1300. lation of gene expression in the hybrid cells. 20. DeFrancesco, L. & Attardi, G. (1980) J. Mol. Biol., in press. Selective suppression of gene expression at other loci has been 21. Eliceiri, G. L. & Green, H. (1969) J. Mol. Biol. 41, 253-260. previously found in mouse-human hybrid cells. Thus, the nu- 22. Perry, R. P., Kelley, D. E., Schibler, U., Huebner, K. & Croce, clear rRNA genes of a single species are transcribed in C. M. (1979) J. Cell. Physiol. 98,553-560. (7, 21) 23. Huebner, K., Shander, M. & Croce, C. M. (1977) Cell 11, 25- spite of the presence of both sets of rRNA genes in the hybrid 33. cells, as shown by genomic blots (22). In addition, the functions 24. Marshall, C. J., Handmaker, S. D. & Bramwell, M. E. (1975) required for the replication of species-specific viruses that are J. Cell Sci. 17, 307-325. Downloaded by guest on September 24, 2021